巴比特论坛

发表于 2017-9-21 10:47:57 | 显示全部楼层
活动类型:
AMA
开始时间:
2017-9-22 10:00 至 2017-9-22 12:00 商定
活动地点:
第48期
性别:
不限
已报名人数:
0

本期嘉宾:
Elizabeth Stark
闪电实验室(Lightning Labs)联合创始人兼CEO,同时为AI及区块链初创公司提供咨询服务,在此之前于斯坦福大学机耶鲁大学任教科技及互联网相关课程,她担任Thiel奖学金的导师,与Google和Mozilla等公司合作,并且是斯坦福大学StartX的企业家。Stark毕业于哈佛法学院。


Rusty Russell
Blockstream基础架构工程师,澳大利亚的自由软件程序员和倡导者,以他在Linux内核的网络子系统和文件系统层次标准方面的工作而闻名。Rusty Russell起草了闪电网络协议规范的大部分内容。


Olaoluwa Osuntokun
闪电实验室(Lightning Labs)创始人,比特币工程师,比特币、分布式系统、应用密码学。这位93年的男生还在加州大学圣芭芭拉分校攻读计算机科学专业。


隔离见证已在比特币上激活,这是改善比特币的可用性和可扩展性的进程中的重要一步。


早前,闪电网络团队已宣布了其在实现各种闪电网络规范部署方案之间的兼容性方面做出的努力。目前团队已经开始测试三种部署方案之间的互操作性:c-lightning,eclair和lnd。同时还要宣布相关的交叉部署测试网站的发布:https://cdecker.github.io/lightning-integration

闪电网络部署方案的开发工作仍在进行中,并会在主网发布任何软件之前进行进一步的测试。欢迎任何想要参与进来为测试和开发工作做出贡献的小伙伴,共同构建第二层可扩展性技术的未来!

关于闪电网络的一切问题,欢迎来问!


胖小喵的论坛小号在打假
您需要登录后才可以发帖 登录 | 立即注册 | 用新浪微博登录

最新最热
hellobitcoin 2017-9-21 13:18:15
闪电网络目前进展如何?能带来些什么实际作用?
预计何时登陆主网?
闪电网络实验室目前有几位开发者呢?你们收入来源是什么?

回复 收起
2017-9-22 10:29:00 ElizabethStark: 我一直认为LN能够实现我们一直以来对比特币的愿景,即能够使用各种应用或设备无缝地发送资金,而不必信任任何中心化机构。LN不仅能使比特币可以以任何数量进行大规模即时交易(包括非常小的价值),而且能够使用户在比特币和莱特币等货币之间进行无信任的兑换。
I’ve always seen LN as fulfilling the vision that we always had for bitcoin: being able to send money seamlessly within a variety of apps or devices, without having to trust a central party. LN will not only bring instant transacting at scale to bitcoin at any amount, including small values, but it can also enable users to trustlessly swap between currencies, such as BTC and LTC.

最近我们有一个团队退出了闪电实验室,而且我们刚确定我们公司的核心价值之一是“don’t yolo security”。基本上,我们想要非常认真地对待安全问题,因此,我们认为,在我们的软件在比特币(或莱特币)主网上线之前,对其进行彻底的测试是非常重要的。目前,我们并没有在发布日期上做出任何承诺,但我们正在努力地工作!
We recently had a team retreat for Lightning Labs where we decided one of our core company values is “don’t yolo security.” Basically, we want to take security very seriously and as a result, we believe it’s important to test our software thoroughly before putting it on the bitcoin (or litecoin) mainnet. For now, we’re not making any promises on dates but we’re working as fast as we can!

有各种各样的公司和团队正在开发闪电网络协议软件。我们公司,闪电实验室,现在共有6名员工,我们在Github上LND的repo还有许多其他开源贡献者(github.com/lighntningnetwork/lnd)。Blockstream至少有两名工程师正在开发C-Lighting部署,一家名为Acinq的法国初创公司有一些工程师正在开发Elair,还有来自麻省理工学院DCI的Lit项目。
There are various companies and teams working on Lightning Network protocol software. Our company, Lightning Labs, has a total of 6 people working for us right now, and we have numerous other open source contributors to our github repo for LND (github.com/lighntningnetwork/lnd). Blockstream at least two engineers working on the c-lightning implementation, a French startup named Acinq has a few engineers working on Eclair, and there’s the Lit project from MIT DCI among others.

目前我们是有天使投资人支持的小型初创公司,但最终我们计划是通过创建工具和服务来盈利,从而让人们更好的使用LN。闪电协议将始终保持开源!
Right now we are a small startup with angel investors, but eventually we plan on making money from building tools and services to better enable people to use LN. The protocol will always remain open source!
2017-9-22 10:17:06 RustyRussell: Hi!  There are three teams working on implementations, and we’ve finalized the specification; we’re working on making sure we’re fully compatible.  It will give bitcoin instant, small transactions.
嗨!现在有三队人正在准备闪电网络的部署,而且我们已经完成的闪电网络规范。我们现在
正在努力确保我们能完全兼容。它会为比特币提供即时的小额交易。

Good question; nobody wants to release before we’ve made sure we’re all compatible.  We’ll know when we’re finished, but it’s likely weeks.  I’m not sure which team will be first to release a mainnet alpha release; who is bravest? :)
好问题;没有人想要在我们确定我们是完全兼容的之前进行发布。我们会知道我们时候大概能完成,但可能需要几周。我不确定哪个团队会首先发布一个主网的alpha版本;谁会打头阵呢?:)

收起
Dearlife 2017-9-21 13:54:01
闪电网络什么时候能发布正式版?
有了闪电网络,比特币能够在小额消费中使用吗?
回复 收起
2017-9-22 10:45:15 ElizabethStark: 从我们的角度来说,测试是最重要的。幸运的是,我们的LND部署在slack频道上有一个充满激情的社区,而且我们进行的测试越多,越能使协议保持稳定,我们就能越快发布可使用真正的货币的测试版。
From our standpoint testing is of utmost importance. We’re very lucky to have a passionate community of supporters on our slack channel for our LND implementation, and the more we can test and make the protocol stable, the sooner we can release it as a beta available for use with real money.
2017-9-22 10:40:30 RustyRussell: 我们马上就要完成了。我们对这个规范很满意,但是之后闪电实验室的Laolu发现了一个我们必须要解决的小漏洞。一旦所有人都对他们的部署满意,而且我们能进行互操作之后,每个团队都会发布他们各自的部署。然后我们就能确定规范是正确的,然后将其标为1.0版(然后开始起草1.1版,我们还有很多更酷的事情要做!)
We’re close; we’re pretty happy with the specification, but then Laolu from Lightning Labs found a small bug which we have to all fix.  Once everyone is happy with their implementations, and we all interoperate, each team will do their own independent releases.  Then we’re sure the specification is right, so we’ll mark it 1.0 (and start working on 1.1, since there are so many more cool things to do!).

是的!不像比特币网络那样按照重量收费(交易的大小)因为这是添加一笔交易的代价,闪电网络是根据发送的金额来收费(例如,0.5%),因为这是发送交易的成本。这意味着小额交易在闪电网络上更有意义。
Yes!  Unlike the bitcoin network which charges by weight (the size of the transaction) because that’s what it costs to add a transaction, Lightning charges by amount (eg 0.5%) since that’s what it costs to forward a transaction.  This means tiny payments make much more sense on Lightning.
收起
dailycafe 2017-9-21 14:29:18
你们会在BCC网络部署闪电网络吗?
如果11月比特币再次硬分叉,你们会选择部署在哪一个网络?或者支持哪一个网络?

回复 收起
2017-9-22 10:49:04 RustyRussell: 在我知道的人里面,没人会这么做。目前的规范充分利用了隔离见证:如果要在BCC上部署闪电网络,我们需要对规范进行彻底更改。我们用了18个月时间才有了隔离见证,而且有很多人参与其中;我不知道到由谁愿意为bcash花费这么多的时间。
Nobody I know plans to do this.  The current specification makes full use of segregated witness; it would need to be completely changed and made worse.  It took 18 months to get this far with the Segregated Witness implementations with multiple people working on it; I don’t know that anyone wants to spend that time for bcash.

我不认为会硬分叉。但如果真的分叉了,我会支持主链,而不是B2X链,但我不能代表其他人。在没有双向重放保护的情况下,网络分裂会非常麻烦。交易可能从一条链泄露到另一条链,所以分裂期间通道无法正常工作。
I don’t think it will fork.  But if it does, I will be supporting the main chain, not the B2X chain, but I can’t speak for the others.  They are making the split very difficult without two-way replay protection; transactions can leak from one chain to another so channels can’t work well during the split.
收起
加油你是最胖的 2017-9-21 14:31:11
据我所知,闪电网络有很多不同的版本,可以介绍一下你们这一版本的特点吗?
目前各个不同的闪电网络是否已经实现互通?未来是否有标准化闪电网络的计划?感觉有太多版本不同的闪电网络会不会太乱了?
回复 收起
2017-9-22 11:01:52 RustyRussell: 我们对我们做事的方式达成了共识(这里是完整的规范https://github.com/lightningnetwork/lightning-rfc )。这对用户来说显然是对好的选择,但这也意味着我们找了尽可能多的专家来做设计方面的决策。每个团队都对闪电进行了改进。我们部署的GUI还不够好,所以我们有点嫉妒其他两个团队 :)
We all agreed on how we’d do things (there’s a complete specification https://github.com/lightningnetwork/lightning-rfc ).  That’s obviously best for users, but it also means that we got as many experts as possible making the design decisions; each of the teams brought improvements to Lightning.  Our implementation doesn’t have a nice GUI, so we’re a bit jealous of the other two :)

是的,我们已经实现了互通(除了我们现在正在清理的漏洞)。https://github.com/cdecker/lightning-integration 上面有测试,但是我们或许我们应该加入一个好一点的测试矩阵,这样可以方便你查看进度,而不用问Christian我们做到哪一步了。
Yes, we’re all synced (except for bugs, which we’re squashing now).  https://github.com/cdecker/lightning-integration has the tests, but we really should add a nice test matrix so you can follow progress without having to ask Christian how we’re going.
收起
王雯敏Wendy 2017-9-21 14:36:12
除了作为比特币第二层扩展技术以外,闪电网络在其他领域是否有应用的可能?
回复 收起
2017-9-22 12:19:22 MaxFang: 你可以想象,在比特币本身就做的很好的地方,闪电网络会更棒。您可以利用比特币不需要个人信息来访问这个全球金融网络的特点,因此任何你可以从这个开放的集成中受益的地方都是非常棒的。如果物联网成为现实,那么闪电会是一种在设备之间转移价值的好方法。想象一下,所有的一切都可以及时到账,从支付工资到通过微支付付费墙对在线内容付费。小额支付使你可以向世界上的任何人付款,而你只需要知道他的闪电地址。如果你使用闪电进行日常支付,这样你经常光顾的那家咖啡馆就不用担心你先用比特币进行支付,然后立即将同一笔钱用于另一笔0-conf交易。
You can imagine that Lightning will be great wherever Bitcoin itself tends to excel. You could leverage how Bitcoin doesn't require personal information to gain access to this global financial network, so anywhere you could benefit from this open integration would be great. If IoT becomes a thing, Lightning would be a great way to transfer value between devices. Imagine instant settlement for everything, from paying wages by the minute to monetizing online content through micropayment paywalls. Micropayments that don't reside within the walled garden of a bank enables you to pay anyone in the world where all you know is their Lightning address. Using Lightning for every day payments, so that your coffee shop doesn't have to worry about you making a Bitcoin transaction and immediately double-spending it with a competing 0-conf transaction.
2017-9-22 12:17:24 ElizabethStark: 我们把LN当做是比特币应用开发的平台。对于LND来说,现在我们有一个建立在我们的部署之上的蓬勃发展的开发人员生态系统,而且我相信来自闪电实验室的Max对此会有更多的发言权。作为开发人员,当你可以进行及时,可扩展的交易时,构建应用程序的体验会刚好。在这里你可以看到一些早期的应用程序的例子:http://lightning.community/software/lnd/lightning/2017/07/05/emerging-lightning-developer-ecosystem/
We view LN as a platform for app development on bitcoin. For LND we have a now thriving ecosystem of developers building on top of our implementation, and I’m sure Max from Lightning Labs will have more to say about this. It a much better experience to build apps when you can have instant, scalable transactions as a developer. You can see some examples of early apps here: http://lightning.community/software/lnd/lightning/2017/07/05/emerging-lightning-developer-ecosystem/
2017-9-22 11:37:25 RustyRussell: 在哪儿都能使用比特币。即时支付。小额支付。我们还有很多关于这个能做什么的想法,我都不知道选哪个是我最喜欢的!我确定你肯定也有很多想法。
Using bitcoin everywhere.  Instant payments.  Tiny payments.  There are so many ideas on what this could do that it’s hard to pick a favorite!   I’m sure you can think of many, too.
收起
alphlo 2017-9-21 14:38:06
你们会如何避免闪电网络中心化的问题,通过什么样的方法呢
回复 收起
2017-9-22 12:48:24 LaoluOsuntokun: 通常很多人把中心化当成是二进制的(0或1),而不是一个梯度的指标。所有去中心化的系统都有中心化的问题,比如,在很多情况下,它可以让我们绕过一些非常困难的问题,而不必考虑各种系统设计的权衡。幸运的是,由于对“超大节点”的形成的约束、准入门槛低,以及用户几乎不存在的切换成本,我预计通道图本身会是相当分散的。
Typically many people speak of centralization as if it were on a binary scale (0 or 1), rather than a sliding gradient. All decentralized systems have centralization pressures, as in many cases, it allows one to sidestep very hard problems and not have to consider various system design tradeoffs. Thankfully due to the disincentives against the formation of large “mega-nodes”, the low barrier to entry, and the near non-existent switching costs for users, I anticipate the channel graph itself will remain fairly diffused.

一个可能的场景是:仅使用单个节点的通道的用户可能会发现,在节点被目标节点断开的情况下,他们的通道有时候可能无法进行更新。这样,用户会有动机在通道图的不同区域内维护多个通道,以确保他们的通道可以进行更新,同时为了保留容错特性(否则,一个节点离线就会使系统无法使用)。
Tackling a possible scenario head on: users that maintain channels with only a single node may find that at times their channels may not be available for updates in the case of node flapping by the target node. Users are incentive to maintain several channels within distinct areas of the channel graph in order to ensure high availability of their channels for updates, and also in order to retain fault tolerant properties (as other wise a single node being offline renders the system unusable for them).

一个拥有许多通道的大型节点不一定比拥有多种通道的多个独立节点盈利能力更强。这是因为一个闲置的通道从所讨论的节点的角度来看是不盈利的。高速通道(那些频繁更新的通道)会更容易盈利,因为费用是在支付完成后支付的。因此,寻求利润的节点会利用新的流动性匹配服务以最大程度的利用潜在的未使用的比特币,而不是把它们的比特币分配给一个“超大节点”。
A large node with many channels my not necessarily be more profitable than a number of independent nodes with a diverse set of channels. This is due to the fact that an idle channel isn’t a profitable one from the perspective of the node in question. High velocity channels (those that have many frequent updates) will be more profitable as fees are paid upon payment completion. As a result, profit seeking nodes will seek to utilize emerging liquidity match making services in order to maximize the latent Bitcoin rather than attempting to allocate their coins to a single “mega node”.
2017-9-22 11:40:03 RustyRussell: 好问题!我们会让运行节点变得尽可能简单,因为你需要运行节点来接受或发送付款。我们希望闪电会有许多节点。你不需要什么特殊的设备,所以每个人都应该运行一个节点。同时,该软件会在做选择的过程中变得更好,避免中心化的问题:我们会有多条通道连接到网络的不同部分,并尝试使用网络上的不同路径进行支付。任何试图搞一个特别大的节点的人都会遇到可扩展性的问题,而且需要在闪电中预存比他们需要的更多的资金,这对他们来说不是很有用。
Excellent question!  We can make it as easy as possible to run a node, and since you need to run a node to accept or make payments, we expect many nodes.  You don’t need special equipment, so everyone should do it.  The software will get better at making choices which avoid centralization: have more than one channel connecting to different parts of the network, and trying to use different paths over the network for payments.  Anyone who tries to create a very large node will have scalability problems and also will have more funds in lightning than they need, which won’t be very useful for them.
收起
gg 2017-9-21 14:39:11
闪电网络部署方案的开发工作过程中 遇到过哪些坑?
回复 收起
2017-9-22 13:05:34 RustyRussell: 把用户吓跑。比特币用户很喜欢尝鲜,即使我们还没有准备好:)我们理解这是很令人兴奋的,但是亏钱就不是什么好事了,所以我们想确保大家已经很清楚的了解了其中涉及的风险。软件从来不是完美的。
Scaring off users!  Bitcoin users love to try stuff, even if it’s not ready :)  We understand, it’s very exciting.  But it’s not exciting to lose money, so we want to make sure people understand the risks very clearly.  Software is never perfect!
收起
okpl6111 2017-9-21 14:54:34
侧链能干什么?为什么说侧链可能会导致中心化?哪些人可以部署侧链?
回复 收起
2017-9-22 13:07:56 RustyRussell: 侧链和闪电是非常不同的。在目前的侧链中,一些节点会保存你在侧链上的比特币。每个人都可以搞一个侧链。详见http://elementsproject.org/ 。这些节点只有在大部分的节点是诚实的时候才安全,这点和比特币或者闪电截然不同。然而,侧链可以让你体验一下区块链,这与比特币的区块链非常不同,但使用的是真正的比特币。此外,还有一些尚未实施的去信任的侧链计划,因为有很多细节需要解决。
Sidechains are very different from lightning; in a sidechain today, a number of nodes hold your bitcoin while it is on the sidechain.  Anyone could set one up; see http://elementsproject.org/ for an example. They are only secure if most of those nodes are honest, which is very different from bitcoin or lightning.  However, sidechains let you experiment with blockchains which are very different from bitcoin’s blockchain, but using real bitcoins.  There are also plans for trustless sidechains which hasn’t been implemented yet, as there are many details to work out.
收起
Aitlas 2017-9-21 15:00:38
本帖最后由 Aitlas 于 2017-9-21 15:03 编辑

1.As I understand, after opening a new payment channel, users need to use their privite keys of that 2 of 2 funding address to sign later commitment transactions online. How could devs mitigate the risk of exposing keys?

2.Frow what I have read from some posts on reddit(shame that bitcointalk is blocked here), there would be two amount limits for the channel. One is the amount of BTC that users cound charge when opening it.
The other is the total amount of BTC that the channel could receive. Someone testing eclair version told me that the former one is 0.042BTC fer each side and the latter one is around 0.15BTC.
Can you confirm it? Besides,do those small amount limits (if they do exist) relate to the exposing risk I mentioned above?



回复 收起
2017-9-22 13:42:25 LaoluOsuntokun: 我们可以做的一件事是尽可能减少攻击者在资金密钥(2-of-2 multi-sig)密钥泄露的情况下可以采取的一系列行动。防御的一个要点是,在攻击者尝试关闭合作渠道的情况下,防止给定地址被随意关闭。如果我们预先提交一个送达地址(在通道关闭的情况下将资金发送到该地址),那么攻击者只能关闭到该通道所有者的地址的通道。预先提交的地址甚至可以让私钥离线,这意味着已经受到攻击节点不会允许攻击者在任何情况下使用这些资金。
One thing we can do is minimize the set of possible actions an attacker can take in the case of funding key (2-of-2 multi-sig) key compromised. A point of defense is guarding against arbitrary closing to a given address in the case of an attempted cooperative channel closure by the attacker. If we pre-commit to a delivery address (where the funds are sent to in the case of a channel closure), then the attacker can only close out the channel to the address of the channel’s owner. The address pre-committed to could even have it’s private key offline, meaning compromise of the node wouldn’t allow the attacker to spend those funds in any case.

更智能的签名过程甚至专用硬件还可以帮助减轻攻击向量,因为服务或企业可能需要第二级授权才能真正触发一个承诺更新(更新通道的余额)。在这种情况下,未经授权的承诺更新将被拒绝。对于在公共/私有子图中运行多个节点的企业,“边界节点”也可以作为额外的防御层,并拒绝任何未经授权的转出。然后,攻击者将需要破坏所有这些节点(甚至不假设硬件签名),以便将资金转移到子图之外(可能是对他们自己)。
Smarter signing processes or even dedicated hardware can also help to mitigate attack vectors as a service or business may have a second level of authorization required to actually trigger a commitment update (updating the balances of channels). In this case, an attempted unauthorized commitment update would itself be rejected. In the case of business running several nodes in a public/private sub-graph, “border nodes” could also serve as an additional layer of defense and reject any outgoing unauthorized transfers. The attacker would then need to compromise all of these nodes (even not assuming hardware signing) in order to transfer funds outside of the sub-graph (possibly to themselves).

另一种解决方法是(目前建议)是创建一组不同的密钥,这些密钥是HTLC签署新的承诺状态所必需的。进一步隔离这些密钥的处理和存储可以保证攻击者在不知道所有密钥的情况下无法真正创建新的状态。(这些密钥本身可能位于不同的机器上)。
An additional mitigation (currently suggested), would be to create a distinct set of keys that are required to sign the HTLC’s of new commitment states. Further isolating the handling and storage of these keys would ensure that an attacker is unable to actually create new states without full knowledge of all the necessary keys (which themselves may be on distinct machines).
2017-9-22 13:03:43 RustyRussell: 好问题!将你的密钥放到网上是闪电的一个弱点:它不是冷储存的。但就在几个月前,我意识到,任何让比特币变得更有用的东西都是这样的,不仅仅是闪电!如果你要使用你的比特币,你需要使用你的私钥。实际上,第一个版本(不仅仅是ECLARI)中的限制是为了避免有人过于热情而将数千美元投入到早期开发的软件中的问题。另外,大型支付在主链上其实更便宜(也更可靠)。
Excellent questions!  Having your keys online is a weakness of lightning: it’s not cold storage.  But a few months ago I realized that this is true for anything which makes bitcoin more useful, not just lightning!  If you’re going to use your bitcoin, you need to use your private keys. And indeed, the limits in the first versions (not just ECLAIR) are designed to avoid the problem of someone getting too enthusiastic and putting thousands of dollars into early-development software.  Plus, really large transfers are cheaper (and more reliable) on the main chain.
收起
starfei 2017-9-21 15:34:01
您好:
想咨询以下几个问题:
@Elizabeth Stark
1.请问您在与初创团队接触过程中,会通过哪些因素来判断团队&项目是否靠谱?
2.另外,国外区块链初创团队可以获得哪些帮助(比如来自FA的办公场地支持、政府补贴、社区开发者支持等)?
3.斯坦福/耶鲁大学是否有准备开设区块链相关课程?中国此前有开展过区块链大学行的活动,形式更多是嘉宾演讲为主。
@Rusty Russell
1.作为自由软件程序员,您如何安排自己一天的生活?
2.在起草闪电网络协议的过程中,哪条协议最让您头疼?
@Olaoluwa Osuntokun
1.创办闪电网络实验室的初衷是什么?今年内预计达成怎样的目标?
2.如果让您用一句话来向朋友介绍现在在做的事(我在和朋友解释区块链相关工作事,常常感到无从说起),你会怎么说?

希望能听到嘉宾的回答~非常感谢巴比特社区的组织!
回复 收起
2017-9-22 13:38:54 ElizabethStark: 1. 我觉得很多人都会从ICO中学到一些非常有趣的经验。:)
I think a lot of people are going to learn some really interesting lessons about this in the ICO space. :)

我比较喜欢那些从产品或技术的角度上非常有潜力,但不鼓吹他们的能力的团队。我相信人们应该设立远大的目标,但不应该夸大其词。 我寻找的是那些超级聪明,热爱自己正在做的事情,但同时也是诚实的,不自欺欺人的团队。
I look for teams that are doing things that have great promise from a product or technical standpoint, but without overselling their capabilities. I believe people should be shooting for the stars, but not selling snake oil. I look for teams who are super smart and passionate about what they are doing, but are also honest and not fooling themselves.

2. 我们还没有收到任何政府资助或帮助,但是比特币社区给了我们很多支持!比如说,我们刚开始的时候,一些朋友慷慨提供了几个月的共用工作空间来给我们办公。我们也很感激所有做测试的以及社区成员为闪电网络的进步和完善做出的贡献。
We haven’t gotten any government aid or funding, but we have gotten a lot of support from the community! For example, when we first started out, we were very lucky to be given free coworking space for a few months from friends of ours. And we are very grateful to all of the testers and community members that have contributed to and helped improve our LND software.

如果有人想要帮助闪电网络软件做测试、做贡献或者在闪电网络上开发应用的,可以查看我们的开发者页面http://dev.lightning.community。
For anyone who is interested in testing, contributing to, or developing apps on our LND software, they can check out our dev website at http://dev.lightning.community.

3.我已经离开学术圈子,进入创业公司的圈子里来了,不过例如斯坦福有一个很棒的关于比特币的课程,由Dan Boneh教授带领,也有讲到闪电网络,你可以在这里看到课程信息:https://crypto.stanford.edu/cs251/
I’ve left the academic world for the startup world these days, but Stanford for example has an excellent course on Bitcoin run by Dan Boneh that has covered Lightning Network in the past. You can check out the course info here: https://crypto.stanford.edu/cs251/
2017-9-22 13:14:26 RustyRussell: 1. 我其实是在Blockstream工作的,但我在澳大利亚远程办公,老板对我的要求没别的,好好开发闪电网络就行了 :)。大多数时候我早上5:15起床,和我在美国及欧洲的同事开电话会议,刷刷推特和Reddit,看一下GitHub上闪电网络及C-闪电网络部署相关的issue和pull request。我的家人大约7点起床。我大约9点回到一个很棒的共用工作空间http://www.base64.com.au/去工作。我会喝很多美味的咖啡,如果高效完成工作的话,会把我的补丁都推到GitHub上,但后就可以休息了。在我睡觉的时候,我在瑞士苏黎世的同事Christian Decker会审阅我的工作。第二天早上醒来就重复这个过程 :)
Well, I work for Blockstream, but I work from Australia and they don’t tell me what to do except keep developing lightning :)  I get up most days at 5:15am to have conference calls with my US and European colleagues, read twitter and reddit and all the github issues and pull requests for the lightning spec and our c-lightning implementation.  My family gets up about 7am, and I work from a really nice http://www.base64.com.au/ shared office from about 9am.  Lots of good coffee, and if I’m productive I push my patches to github, where Christian Decker in Zurich reviews them while I’m sleeping. Then I do it again :)

2. 有两个复杂的部分是我没有料到的。第一个是向两个不同的方向发出许多非常快的交易。我之前做了一个安装,但对于闪电实验室来说太慢了,所以我们又设计了一个更好的协议,不过这个协议没有之前的那个简洁。第二个是如何保证我们能够将惩罚交易安全地外包出去。Tadge Dryja和我一起工作,保证你能够将一组惩罚交易发送到一个“瞭望塔”节点,而该节点不会知道任何关于你的通道的信息。这就意味着如果通道另一端想要作弊的话,你不需要亲自去抓住他(可能你当时不在线),因为瞭望塔可以帮你抓他,这就容易得多了。当时Tadge认为瞭望塔不应该知道他们在帮哪些通道看守,除非出现作弊行为。当时我认为这不可能实现,但最终证明Tadge的想法是对的。所以我们保证我们的交易是以这样的机制进行的,现在Tadge正在开发瞭望塔。
There are two complex parts which I didn’t expect. The first one is making many really fast payments in both directions at once; I did an early implementation which was too slow for Lightning Labs, so we designed a much better protocol which unfortunately was not as simple. The second one was making sure we can outsource penalty transactions safely: Tadge Dryja and I made sure that you can send a set of penalty transactions to a “WatchTower” node without the watchtower knowing anything about your channel.  This means that if the other side of the channel tries to cheat, you don’t have to catch them yourself (maybe you are offline?) because the watchtower can do it.  This would be much easier, except Tadge was determined that watchtowers should not know what channels they’re watching unless there’s a cheating attempt.  I didn’t think it was possible,  but Tadge was right, and it is.  So we made sure our transactions work with this; Tadge is working on WatchTowers now.
2017-9-22 12:41:41 LaoluOsuntokun: 1. 初衷是成立一个实体公司,保证闪电网络的研究、开发以及最终的部署都能有条不紊地进行。此外,我们认为这个行业里缺乏致力于比特币基础设置建设的公司,会导致将来没有足够的具备专业知识的开发者来推动比特币不断发展。我很感激能够组织这样一个团队,我们拥有非常了不起的工程师,大家都怀抱着推动比特币基础设施发展的愿景,包括开发新的闪电网络客户端运行模式、闪电网络软件、改进核心的比特币协议以及多种不同的密码学分支。
The original purpose was to create an entity to see through the research, development, and eventually deployment of the Lightning Network itself. Additionally, we felt that our industry lacked a set of companies dedicated to Bitcoin infrastructure development, leading to a lack of knowledgeable developers who possess the necessary knowledge to keep Bitcoin moving forward in the future. I’m thankful to have assembled such an amazing team of engineers as we have now, who share our vision of advancing the state of the art of Bitcoin related infrastructure spanning from new light client operating modes, lightning network software, improvements to the core Bitcoin protocol, and various divisions of applied cryptography.

今年剩下的日子里,一个重要的目标是进一步开发闪电网络,并且对我们最初发布的闪电网络beta版本作出终稿,这将标志着闪电网络正式在主网上部署。此外我也在努力完善新的闪电网络客户端运行模式中微子(https://github.com/Roasbeef/bips/blob/master/gcs_light_client.mediawiki)。最后,我们还在考虑一些新的方法,在比特币上实现更多更先进的智能合约功能,努力让闪电网络整合到比特币生态圈中。此外,年底前我们希望能够发布新的更高效的支付渠道设计方案。
A major goal for the remainder of this year is to further the development of, and finalize our initial beta release of lnd, which would mark the official mainnet deployment for our software. I’m also very keen on finalizing our new light client operating mode neutrino (https://github.com/Roasbeef/bips/blob/master/gcs_light_client.mediawiki). Finally, we’ve been cooking up a number of new ideas for more advanced smart contract functionality on bitcoin, working on Lightning integration into the Bitcoin ecosystem, and also have newer more efficient payment channel designs which we hope to publish/release by the end of the year.

2. 这个问题好难!很难用一句话来解释,大概就是:我在给一个去中心化的货币系统开发新的私密的、可审计的、抵抗力强的金融设施。
Hmm, that’s a tough one! Hard to do in one sentence, but how about: architecting new private, auditable, resilient financial infrastructure for a decentralized monetary system.
收起
IMJENNIM 2017-9-21 16:07:54
想问一下闪电网络部署不会降低矿工的收益吗?
回复 收起
2017-9-22 11:46:54 RustyRussell: 这个问题提得好!闪电网络有三个影响:1. 部分交易将会在闪电网络上,而不是在主链上。2. 比特币会更加强大。3. 你必须要使用主链才能使用闪电网络。
Very important question! There are three effects here: one, some transactions will be on the lightning network, and not on the main chain.  Two, bitcoin becomes much more powerful.  Three, you need to use the main chain to use lightning.

1. 闪电网络用在小额交易、线上交易最好,因为闪电网络是按照每中本聪收费的(大约在0.1%或0.5%),因为它的成本就这么多;而比特币是按照交易大小或者说重量收费的,因为它的成本与交易大小有关。所以大额交易用主链更划算。此外,你需要在闪电网络里有一个通道通向你要发送的那个地址,这就意味着非常大额闪电网络交易一定会失败。最后,接收转账的闪电网络节点必须在线,而在主链上的交易可以晚点再接收也没关系。
1. Lightning is best for small transactions, online transactions: lightning nodes charge a per-satoshi fee (0.5% maybe 0.1%) because that’s how much it costs them, whereas bitcoin nodes charge by transaction weight (size) because that’s how much it costs them.  So large amounts are cheaper on the main chain.  Also, you need a path through lightning to your destination where each step can transfer the amount: this means really large lightning payments will just fail anyway.  Finally, the receiving lightning node needs to be online, whereas with the main chain they can collect the money later.

2. 闪电网络使得比特币更强大。我们可以支付1毛钱、1分钱这种小额交易,并且能做到瞬时转账,这就打开了一个新世界的大门,没有任何其它支付方式能够做到。闪电网络实现后,比特币价格会比现在还要高。
2. But lightning makes bitcoin more powerful.  We can pay 0.1 CNY!  0.01 CNY!  We can do instant transactions!  And that opens a world which no other payment method can reach.  Whatever bitcoin is worth now, it’s worth much more with lightning added.

3. 要在闪电网络上开通一个通道,你需要一个在链上的交易,就和再往通道里充更多钱或者另一端无法达到的话就关掉差不多。目前我们一周能够开通一百万个通道,这可能听起来很多,但实际上要开通十亿个的话需要花上20年!在可预见的将来,如果闪电网络流行开来,链上交易的需求会增加,那么用户就会付给矿工一份相当不错的奖励。
3. To open a channel in lightning, you need a transaction on chain; similarly to add more total funds to a channel, or close it if the other side is unreachable.  We can open about 1 million channels a week at the moment, which sounds like a lot until you realize that it will take 20 years to get 1 billion channels!  For the forseeable future, if lightning becomes popular it will increase the demand for onchain transactions, and so the users will be paying miners a very nice reward!
收起
ICOCHINA 2017-9-21 18:31:14
请问
1、闪电网络如果不依赖Segwit,是否有其他实现的方式?是不是只要用任何方法解决交易可锻性问题,都可以上闪电网络?
2、闪电网络会导致中心化吗?很多人说,闪电网络的节点容易被政府控制,进而导致比特币的中心化。
3、闪电的开关通道还是要用主链交易,闪电只能缓解拥堵,最后打开和关闭通道的人多了,如果1M锁死,还是一样会堵。对吗?这样是不是应该主链扩容?
4、闪电网络和侧链有什么不同?
谢谢


回复 收起
2017-9-22 12:11:53 ElizabethStark: 1. 发布不带可锻性修复方案的闪电网络,就相当于发布劣质的、没什么用处也不安全的软件。理论上讲可以,但是并不理想。此外,做技术开发的当然想要把这项技术做到最好。今年夏天隔离见证在比特比网络上激活对于我们来说很重要,我们甚至还开了香槟庆祝:https://twitter.com/lightning/status/888203288675459076!
Releasing LN without a malleability fix is akin to releasing inferior software that doesn’t have good usability and security attributes. It’s possible in theory, but undesirable. Plus, if you’re building technology you want to make it the best it can possibly be. The activation of segwit on bitcoin this summer was a very big deal for us (and we drank some champagne to celebrate: https://twitter.com/lightning/status/888203288675459076)!
Picture1.png
理论上讲还有其他的可锻性修复方案,但隔离见证还带来了容量增加的好处。因为有了隔离见证,我们已经在能比特比网络上看到大于1MB的区块了。https://www.smartbit.com.au/block/484399/transactions?sort=size&dir=desc
In theory there could have been another malleability fix, but segwit had the added benefit of adding a capacity increase instead. We are already seeing blocks larger than 1MB on bitcoin thanks to segwit. https://www.smartbit.com.au/block/484399/transactions?sort=size&dir=desc

2017-9-22 11:57:51 RustyRussell: 1. 你可以开通一些临时(而非永久的)通道,他们不需要交易可锻性保护。此外不用隔离见证的话,如果该通道失败的话花的时间也会更长一些。理论上讲这是可行的,但是比较麻烦。
You can make temporary channels without malleability protection, not permanent ones.  Also the worst-case time if a channel fails must be longer if you don’t have segregated witness. But it’s theoretically possible to do, but a lot of hard work.

2. 每个人都可以运营一个节点:就好像你用自己电脑的CPU进行比特币挖矿一样。要开通一个通道需要投入一点比特币,这是要运营闪电网络节点的唯一入门要求。
Anyone can run a node: it’s more like when bitcoin mining could run on your CPU.  You need to have some bitcoin to open a channel; that’s the main barrier to running a lightning node.

3. 我认为我们最终还是要对主链进行扩容的,现在有了隔离见证之后可以实现更大区块,所以大家都在等着看会发生什么。我们还有一些方法可以让区块容纳更多交易,到时候也会用上这些方法。但我希望闪电网络的用户数量能够达到十亿,所以我们需要更大的区块。这个事情现在还不用着急,但是我们需要花很长时间来制定一个人人满意的方案,所以现在已经有开发者在评估可选的各项措施了,比如说可以在这里查看: https://bitcoinhardforkresearch.github.io/https://bitcoinhardforkresearch.github.io/
I believe we will need to scale up the main chain eventually, and everyone is watching what happens now we have larger blocks with Segregated Witness.  We have a few more ways to squeeze more transactions into each block which I’m sure we’ll do, but I want a billion people using lightning, so we need larger blocks.  There’s no hurry, but it will take a long time to plan and make sure everyone agrees with the plan, so developers are already evaluating what options we have, such as https://bitcoinhardforkresearch.github.io/

4. 侧链是一个使用真正的比特币的区块链,并且可以把比特币在主链和侧链上移动(一般来说,当比特币放在侧链上时,一组节点会储存这些比特币)。使用比特币来在不同的区块链上做实验是很有趣的,但你得信任这些节点才能让他们储存比特币。
A sidechain is a blockchain which uses real bitcoins, and has a way of moving them to and from the main bitcoin network (usually, some group of nodes which hold the bitcoin while they’re on the sidechain).  It’s cool for experimenting with different blockchains but still using bitcoin, but you have to trust those nodes to hold the bitcoin.

闪电网络是多对节点组成的网络,每个节点存有一个已签名的比特币交易,可以随时花掉。使用方式有二:一是不花掉这些比特币交易,节点对可以达成共识对比特币交易进行更新。假设该比特币交易将给你10000中本聪,给我20000中本聪,我想给你8000中本聪,那么我们就把交易更新为给你18000中本聪,给我12000中本聪。二是你可以给该交易一个指令,只有我给另一个人付了款之后,该交易才会给我付款。所以你可以使用这个网络给任何一个人付款,因为只有我按照你的指令做了才能收到你的钱款。无论何时你都知道你手里掌握着最新完成的已签名比特币交易,如果发生了什么意外,你也可以在主链上正常交易。这个机制真的很巧妙,我真希望这是我发明的。
The lightning network is a network of pairs of nodes, each holding a signed bitcoin transaction they can spend at any time.  There are two tricks which make it work: first, instead of spending their bitcoin transaction,  pairs can agree to update it. Say the transaction pays you 10000 satoshi and me 20000 satoshi; if I want to pay you 8000 satoshi, we update it so it pays you 18000 satoshi and me 12000 instead.  Second, you can make the transaction pay me only if I pay someone else!  So now you can make a payment through the network to anyone, knowing you only have to pay if it works. And any time, you know you’re holding the latest complete, signed, bitcoin transaction so if something goes wrong, you spend it on the main chain like normal. It’s really clever; I wish I’d invented it!
收起
币发展 2017-9-21 18:39:08
咨询

1、未来什么样的交易用比特币主链,什么样的交易用闪电网络?
2、闪电网络什么时候能在比特币中,让普通人使用?它能解决拥堵问题吗?
3、小区块+闪电望到尽头的结果,会不会导致手续费很高?闪电网络能解决手续费高的问题吗?
4、闪电网络会用在BCC上吗?
5、闪电网络有可能会用于以太坊吗?
回复 收起
2017-9-22 15:32:34 ElizabethStark: 4. BCH/BCC没有可锻性修复方案,没有可锻性修复的话就没法安装闪电网络。
BCH/BCC doesn’t have a malleability fix, and without one they wouldn’t be able to run any of the active implementations of LN.
2017-9-22 15:30:45 RustyRussell: 4. 我曾经试图运行一个BCC节点来分裂我的币,但是没成功于是就放弃了,把他们转手卖掉了。可能现在好一点了?
I tried running a BCC node to split my coins, but I couldn’t get it to work and gave up: sold them a different way.  Maybe it’s better now?
2017-9-22 15:26:06 MaxFang: 3. 高额的手续费是因为每个区块都在相互竞争空间,因为每个区块只能容纳约1MB的交易。闪电网络可以让你给任何连接到闪电网络的人转账,并且不需要在链上清算每笔交易。所以如果闪电网络上有足够多的整合,那么交易费用将会减少,因为对区块空间的竞争将会变小。最终还是取决于有多少人在使用比特币,以及有百分之几的网络在闪电网络上使用链下交易。
High transaction fees is caused by competition for space within each block, since a block can only hold about 1 MB of transactions. Lightning allows you to make payments to anyone else connected to the Lightning Network without settling each payment on-chain, so if there is enough integration on Lightning, it will help reduce transaction fees, due to there being less competition for space in blocks. In the end, it will also depend on the number of people using Bitcoin, as well as the percentage of the network that is using off-chain payments on Lightning.
2017-9-22 15:01:36 LaoluOsuntokun: 1. 两种交易模式都有其利弊。极端情况下,我认为非常大额的、一次性付清的交易会在链上进行交易,而小额的、频繁的交易则会使用闪电网络。如果有开发者想要把即时转账功能整合到他们的应用中去的话,闪电网络能够提供更简单的应用模型,不需要不断地花费、管理比特币交易输出。
Both modes of transacting have their various tradeoffs. On each extreme side, I envision very large one off payments to take place on-chain, while smaller more frequent payments will end up using Lightning. For developers looking to integrate real-time payments into their applications, Lightning also offers a much simpler application model than that of continually spending/managing Bitcoin outputs.

4. 我没听说过有哪个闪电网络的开发者在研究BCC,甚至没见有人在运行BCC节点。
I don’t know of any LN developers who are actively working on BCC, nor even run BCC nodes.

5. 以太坊同样能够支持支付通道。有一个工作组正在开发一个和闪电网络相似的东西,但是最基本的设计原理和闪电网络在比特币上的应用差别非常大。
ETH can support payment channels as well. There’s a single team (and implementation) working on something similar to Lightning, but the fundamental design is very different to that of LN on Bitcoin.
收起
Bcoin 2017-9-21 18:44:02
1、比特币有闪电网络而以太坊有雷电网络。如今雷电网络的一个简单版本μRaiden已经在以太坊测试网络上发布,μRaiden允许用户相互之间直接进行小额支付。请问闪电网络和雷电网络相比,有什么不同?有优势吗?
2、以太坊现在智能合约在主链上,数据越来越多造成臃肿,理论上来说,智能合约有可能通过硬分叉移动到侧链上吗?
回复 收起
2018-3-6 13:08:48 黄坚强回复LaoluOsuntokun : :lol


补充内容 (2018-3-6 19:12):
test

补充内容 (2018-3-6 19:13):
test
2017-9-22 15:23:10 LaoluOsuntokun: 1. 就我所知,微型雷电网络只是一个基本的以太坊单向交易通道。以太坊没有任何可用的轻型客户端(所以不能实现快速/循环模式同步),所以还是得要有一个全节点(对于全节点可能有不同的定义)。目前多个闪电网络已经在积极开发并在比特币测试网上测试了几个月了。
AFAIK, micro raiden is just an implementation of basic uni-directional payment channels for ethereum. To my knowledge, ETH doesn’t have any deployed functional light client implementations (so not fast/wrap mode syncing), so one would still need a full node (of which they have differing definitions) to interact with. The various LN implementations have been undergoing active development and testing on Bitcoin’s testnet for many months at this point.

要直接比较这两个系统很难。就我所知,雷电网络并不像闪电网络那样有非常全面的规范。一个本质上的区别是雷电网络只能够使用ERC-20代币,而闪电网络则使用底层货币——比特币。我最近一次关注的时候,雷电网络还没有任何要在路由层做改进隐私问题的计划,而闪电网络有基于洋葱路由的Sphinx来改进隐私问题。
It’s hard to directly compare the two systems, as to my knowledge Raiden doesn’t have a comprehensive specification like Lightning does. One distinction (a fundamental one) is that Raiden only works with ERC-20 tokens, while LN uses the base-layer currency: bitcoin. Last I checked, Raiden didn’t have any attempted privacy enhancing features at the routing layer like Lightning does with our Sphinx based onion routing.

要说闪电网络有什么优势,我想应该是我们有一批优秀的、多元化的工程师在开发闪电网络。由于有多个团队,闪电网络的设计和规范需要经过多轮审阅。要我来评判我自己开发的东西,我当然无法给出客观的评价,但我个人认为,相较于雷电网络,闪电网络开发得更完善、可以进行实地部署了。
One advantage I’d say that we have is the shear number of diverse, talented engineers working on LN. With multiple teams, the design and specification themselves go through several rounds of review and consideration before we consider them sound. I’m very biased, but from my PoV Lightning is closer to production deployment than Raiden (the full system).

2. 可以的,他们可以通过硬分叉在系统上开出一个完全有效的侧链(如果我没理解错问题的话?)。
Yes, they could hardfork a fully validated side-chain into the system (if I’m understanding the question?).
收起

本期嘉宾

神回复

AMA时间:2017-09-22 10:00 - 2017-09-22 12:00 热度(9749) 讨论(88)
返回顶部 返回列表

登录

发帖